|Publication number||US8099389 B2|
|Application number||US 12/366,861|
|Publication date||Jan 17, 2012|
|Filing date||Feb 6, 2009|
|Priority date||May 26, 2000|
|Also published as||CA2408924A1, DE60109602D1, DE60109602T2, DE60109631D1, DE60109631T2, DE60112103D1, DE60112103T2, DE60112107D1, DE60112107T2, EP1285517A2, EP1285517B1, US7028025, US7127525, US7490166, US7502784, US8290907, US20020029285, US20020035596, US20020073061, US20030046432, US20060153423, US20090144292, US20100205246, WO2001092973A2, WO2001092973A3, WO2001093525A2, WO2001093525A3|
|Publication number||12366861, 366861, US 8099389 B2, US 8099389B2, US-B2-8099389, US8099389 B2, US8099389B2|
|Original Assignee||Citrix Systems, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (158), Non-Patent Citations (21), Referenced by (1), Classifications (34), Legal Events (2)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This claims priority to and the benefit of U.S. patent application Ser. No. 11/368,235 now U.S. Pat. No. 7,502,784, filed Mar. 3, 2006, that claims priority to U.S. patent application Ser. No. 09/866,531 now U.S. Pat. No. 7,028,025, filed May 25, 2001, that claims priority to U.S. provisional patent application Nos. 60/207,532, filed May 26, 2000, and 60/225,217, filed Aug. 14, 2000, the entirety of which are incorporated herein by reference.
The present invention relates generally to communication between an application server and a thin client in a distributed system and, in particular, to the reduction in the amount of graphical display data transmitted over a low bandwidth transport protocol mechanism.
Distributed computer systems utilize the technique of distributing application execution. More specifically, an application server locally executes an application program and provides the application output data to clients/network users who then display the results on a display screen coupled to their local computer. By performing the application processing functions at the server and displaying the application output at the client, these distributed systems make the best use of server and client resources, i.e., the more capable server performs the compute and memory intensive application processing functions while the client, with potentially greater video performance, displays the output. During execution of the application program, a user of the client views the application output data on the client's display and interacts with the application program via keyboard or mouse inputs. The client user's inputs correspond to requests to the application server to perform certain actions that affect the operation of the application program. The application server passes these requests to the application for further processing. The application server also intercepts the application output data, generated by the application program and typically targeted to the operating system of the application server, and transmits all or part of the data to the client for display. From the perspective of the user, the application program seems to be executing locally even though it is actually being executed on a remote server and just being displayed locally.
The performance of distributed systems depends to a large degree on the capability of the network, which couples the client and the server, and therefore on the amount of graphical data (corresponding to the application output data) that is transmitted over the network. Since the application server and the client workstation are generally linked only by a transport mechanism (such as serial lines, telephone lines, local area networks and wide area networks) that possesses lower bandwidth than would be otherwise available if the application program were executing locally, the perceived performance of the distributed system may suffer. Accordingly, manufacturers of distributed systems/software continually strive to improve the performance of such systems.
The present invention improves the performance of distributed systems by reducing the amount of graphical data (e.g., bitmapped graphical data, encoded bitmaps, glyphs, and line data) transmitted between the server and the client. The invention achieves this reduction in several different ways, for example, by encoding the graphical data into a smaller object, by representing a graphical object with indicia of the object, by increasing the repetitiveness of the data in the protocol stream so that compression algorithms operate more efficiently, by tracking and leveraging the prior transmissions of identical/repetitive graphical objects, by adapting the rate of processing activity or the encoding technique in response to changes in the network performance or in response to performance mismatches between the client and server, and in several other ways described herein.
In one embodiment, the present invention enables the efficient transmission of graphical display data from an application server node to a client node. In this embodiment, the invention intercepts a plurality of graphical display information commands issued by an application executing on the server. Once these commands have been intercepted or otherwise obtained, they are analyzed to determine whether they should be selected for further transmission to remote client nodes. The selected commands are then encoded and packaged into a protocol packet stream for eventual transmission to the remote client nodes. Prior to transmission, the packet stream can be further compressed by subjecting the encoded packets to compression techniques.
The encoding technique of the present invention substantially maximizes compression of the protocol packet stream by building in repeatability into the protocol stream that allows the compression algorithm to substitute pointers for repetitive data. This technique focuses on maximizing the efficiency of the compression algorithm, rather than solely reducing the amount of data via encoding prior to transmission. The result of this encoding scheme is that the degree of compression is significantly increased for many types of graphical data. The encoding technique of the present invention encodes the graphical data so that portions of the graphical data that repeat are encoded in the same manner. The compression algorithm then operates on this repetitive encoded data to realize a greater degree of efficiency.
In one embodiment, the present invention efficiently reduces the amount of repetitive graphical data transmitted to a remote display device coupled to the client. In this embodiment, a server agent searches for an index (that may have been previously stored on the server by the server agent) associated with the graphical data and which is indicative of a prior transmission of the graphical data to a client agent. The index or handle represents a memory location in the client's cache memory that currently stores the previously transmitted graphical data. If the index is found, then the index is transmitted to the client agent, thus avoiding the need to transmit the graphical data itself. The client agent obtains the graphical object stored in the cache location identified by the index and performs additional processing activities as directed by the server agent. In one aspect of the invention, commands and information transmitted between the server agent and the client agent are queued at the server. The commands in the queue can then be processed in accordance with an estimate of the relative performance between the client and server agents or computers.
If the server agent is unable to locate the index, the server agent searches a fuzzy database for a fuzzy key that is associated with the graphical data. The fuzzy key, for example, corresponds to a length of the graphical data. The fuzzy key indicates that the graphical data had at one time been transmitted to the client and that there is a strong likelihood that the graphical data is stored within the client's persistent storage memory. One or more fuzzy keys are provided to the server agent upon initiating a session between the client and server agents. Alternatively, the fuzzy keys are transmitted to the server agent in accordance with a predetermined polling interval. Regardless of how the fuzzy keys are received, the server agent stores them in the fuzzy database for subsequent use during the client/server session. The fuzzy database relating to a particular session can also be maintained beyond the term of the session so that it is available for future sessions, by, for example, having the client agent persistently cache the graphical data from the particular session and causing the server agent to dynamically construct the fuzzy database on the server from the client's persistently-cached graphical data during a future session. This may prove useful in environments where frequent, but short, sessions are present.
If the fuzzy key is located, then the server agent transmits the fuzzy key together with a command to the client agent, which, for example, retrieves the graphical data stored in its persistent storage memory and copies it into its volatile memory. If the client agent fails to retrieve the graphical data, the server agent can transmit the graphical data, together with a recovery marker, to the client agent.
If the server agent is unable to locate the fuzzy key within the fuzzy database, the server agent forms the fuzzy key and stores the fuzzy key in the fuzzy database. The server agent then transmits the fuzzy key, an index, and the graphical data to the client agent, which is instructed to store the data in the client's volatile memory associated with the index. As the client-server session continues and the client's volatile memory becomes full, graphical data will be transferred into the client's persistent storage. In one aspect the client agent stores the graphical data in a file residing in the persistent storage memory that reflects the fuzzy key as part of the file's name.
In one aspect of the invention, the server agent instructs the client agent to update an off-screen surface coupled to the client agent using the graphical data identified by the server agent and/or stored in the client's volatile or persistent storage memory. These instructions can be triggered in response to a change in the off-screen surface, such as when the off-screen surface and on-screen surface of the client node interact. If the client agent fails to update the off-screen surface in accordance with the server agent's instructions, the server agent can transmit the graphical data to the client agent and direct that the client agent update portions of its on-screen or off-screen surfaces, which may have been corrupted, with the transmitted graphical data.
Further, the encoding technique used to encode the graphical data prior to transmission can be selected based on an estimate of the network conditions of the network that couples the client agent with the server agent. This estimate can correspond to a time differential measured by the client agent that relates to the amount of time expended to receive a particular encoded graphical data.
In one embodiment, the present invention adapts a processing rate in response to changing network conditions. In one aspect of the invention, the server agent processes graphical data addressed to the client agent at a first rate. By determining the network conditions of the network that couples the client and server agents, the server agent can adjust its processing rate from the first rate to a second rate in response to a change in the network conditions. The server agent can determine information about the network conditions by transmitting the processed graphical data to the client agent and instructing the client to measure a time differential associated with the transmission or receipt of the graphical data. In this manner, the time differential provides an indicator of the network conditions and the server agent can rationally select the second rate in accordance with this time differential. Similarly, the encoding technique used while processing the graphical data by the server agent at the first rate can be modified to a second encoding scheme/technique in response to changes in the network conditions that are determined by the server agent. The network conditions can be estimated in accordance with the time differential discussed previously. Once the second encoding technique is selected by the server agent, the server agent will process subsequent graphical data using this second encoding technique.
In one embodiment, the invention adapts a processing rate of the server in response to a performance mismatch between the server and the client coupled to the server via the network. In operation, the server agent processes graphical data and determines a first time period associated with such processing. For example, the first time period can be determined by scrolling a frame buffer of the server. The client agent also processes the graphical data and determines a second time period associated with its processing. The server agent then determines the time differential between the first and second time periods and adjusts its processing rate in accordance therewith.
In one embodiment, the invention efficiently reduces the amount of graphical line data transmitted from a server to a client via a communications network. The server agent separates a path defining a line or region into a plurality of strips, where each of the strips have a beginning and an endpoint coordinate defined within a coordinate system (e.g., a Cartesian coordinate system). The coordinate system corresponds to a region of a display surface associated with the client. In one aspect of the invention, the server agent quantizes the coordinate system into a plurality of quantized angles (e.g., 16 quantized angles) and determines the endpoint coordinate of one of the plurality of strips. The coordinate system is then normalized so that its origin coincides with this endpoint coordinate. The beginning coordinate of the adjacent strip in the path is then associated by the server agent with the endpoint coordinate of the first strip. The server agent then selects one of the quantized segments of the coordinate system as corresponding to an approximate angle of the adjacent strip. This technique enables the transmission of the difference between the endpoint coordinates of the two strips (i.e., the strip length) and an indication of the selected quantized angle and thus avoids having to transmit both sets of coordinates or a coordinate and the precise angle for each strip.
In one embodiment of the invention, the server agent remotely controls the formation of an off-screen surface at a client, which is coupled to the server via a communications network. The server agent informs the client agent that an off-screen surface of a particular size or type is to be formed within the client's memory and the client agent responds by selecting and allocating a first memory region having an appropriate format to represent the off-screen surface from a plurality of available memory regions in the client's memory. The server agent then either transmits a bitmap or an indicia of the bitmap (e.g., index or fuzzy key) to the client agent and instructs the client agent to copy the bitmap to a particular location within the first memory region. Once the bitmap is copied to the off-screen surface, it can be used to update the client's on-screen surface. Such off-screen operations can also be performed for non-bitmap data, such as lines or text.
The server agent can also respond to error conditions by storing a duplicate of the client's off-screen surface in the server's local memory. In the event of an error condition, such as when a preselected memory region of the client's memory is not sufficiently large enough to store a particular off-screen surface, the client agent asynchronously informs the server agent of the error. In response to the error condition, the server agent transmits at least one portion of its duplicate off-screen surface or a copy of the bitmap to the client agent and instructs the client agent to update the client's on-screen surface using the transmitted duplicate off-screen surface data.
The foregoing discussion will be understood more readily from the following detailed description of the invention, when taken in conjunction with the accompanying drawings, in which:
The present invention reduces the amount of graphical data transmitted between an application server, which is executing an application program, and a subscribing client, which is locally displaying the output data of the application program, by encoding the data into relatively small representations that repeat within the protocol stream transmitted between the client and the server. In this manner, the invention minimizes the size of each discrete data element that must be transmitted and increases the repeatability of the data within the protocol stream so that compression algorithms that operate more efficiently on repetitive encoded data can realize a greater degree of compression efficiency. The invention uses a number of techniques to realize this enhanced compression, including the following: scanning a command queue for later-issued commands which supercede earlier-issued commands; disk-caching techniques that improve compression of data that has been previously encountered during a client-server session; transmission of relative coordinates using quantized regions to avoid sending both endpoints and/or the angle of an associated strip; and manipulation of off-screen surfaces to enhance performance during the client-server session. The invention also reduces the frequency of overscroll problems encountered when there is a performance mismatch between a fast server and a relatively slow network or client.
With reference to
The client node 110 can be any personal computer (e.g., 286, 386, 486, Pentium, Pentium II, Macintosh computer), Windows-based terminal, Network Computer, wireless device, information appliance, RISC Power PC, X-device, workstation, mini computer, main frame computer or other computing device that has a windows-based desktop and sufficient memory for executing a small, display presentation program. The display presentation program uses commands and data sent to it by the application server 150 across the network 140 to render a graphical display. In a distributed computer system model, the execution of application programs occurs primarily on the application server 150 and the user interface, keystrokes, and mouse movements produced by client-side inputs are transmitted over the network 140 to and from the client node 110.
The client node 110 typically includes a processor 112, a volatile memory 114 (e.g., RAM cache), an operating system 116, a client agent 118, a persistent storage memory 120, a network interface 122 (e.g., a network interface card), a keyboard 124, a mouse 126, and a display 128. Windows-oriented platforms supported by the client node 110 can include, without limitation, WINDOWS 3.x, WINDOWS 95, WINDOWS 98, WINDOWS NT 3.51, WINDOWS NT 4.0, WINDOWS 2000, WINDOWS CE, MAC/OS, Java, and UNIX. The client agent 118 corresponds to a software program that receives commands and data from the server node 150 and from a user (not shown) of the client node 110. The client agent 118 uses the received information when interacting with other components of the client node 110 (e.g., when directing the operating system 116 to output data onto the display 128). The client agent 118 also transmits requests and data to the server node 150 in response to server-issued commands or user actions at the client node 110.
Similarly, the server node 150 includes a processor 152, a volatile memory 154, an operating system 156, an application program 158, a server agent 160, persistent storage memory 162, and a network interface 164. The server agent 160 corresponds to a software program that interfaces with the client agent 118 and other components of the server node 150 to support the remote display and operability of the application program 158.
Each application server 150 hosts one or more application programs 158 that can be accessed by the client nodes 110. Examples of such applications include word processing programs such as MICROSOFT WORD and spreadsheet programs such as MICROSOFT EXCEL, both manufactured by Microsoft Corporation of Redmond, Wash., financial reporting programs, customer registration programs, programs providing technical support information, customer database applications, or application set managers.
During execution of the application program 158, a server 150 communicates with the client node 110 over a transport mechanism (part of the server agent 160). In one embodiment, the transport mechanism provides multiple virtual channels and one of the virtual channels provides a protocol for transmission of graphical screen data from the server node 150 to the client node 110. The server 150 executes a protocol driver (part of the server agent 160) that intercepts graphical display interface commands (generated by the application program 158 and targeted at the server's operating system 156) and translates them into a protocol packet suitable for transmission over the transport mechanism.
In one embodiment and with reference to
More particularly, a protocol driver of the server agent 160 intercepts the graphical display interface commands (step 202). In one embodiment the protocol driver intercepts high-level application programming interface calls made by the application program 158. In one embodiment, the protocol driver intercepts GDI commands provided by WINDOWS-based operating systems. In another embodiment, the protocol driver intercepts QUICKDRAW commands, provided by MAC/OS-based operating systems. In still other embodiments, the protocol driver intercepts lower-level commands, such as interface calls to the device driver associated with the graphical display of the server node 150. In still other embodiments, the graphical display interface command may be duplicated rather than intercepted. In these embodiments, the display of the application program output data on the server 150 and the client 110 will be substantially similar, if not identical. The protocol driver creates a transport protocol packet based on the intercepted graphical display command (step 208). The protocol driver may use a protocol command set that contains an analogous command for each possible intercepted display interface command. Alternatively, the protocol driver may use a command set that is a subset of the intercepted display interface commands. In other embodiments, the protocol command set may be a superset of the possible intercepted display interface commands.
For each intercepted GDI command and affected graphical data that must be transmitted to the client agent 118 in order to display the output of the application program 158 on the client's display 128, the server agent 160 queues the related protocol command together with a unique key 412 (
For example, a series of commands to display the next page of data issued in quick succession need not be displayed individually at the client 110. All that need be displayed is the last display page issued, since this last page of data will overwrite all previous pages of data. By removing the unnecessary commands and unique keys corresponding to the earlier pages from the server's queue, the server agent 160 substantially reduces the amount of data that must be processed and transmitted.
In one illustrative embodiment, only unnecessary commands and related data pertaining to opaque operations are removed from the server's queue. For example, if a previously-queued command involving the creation of a rectangular region having coordinates 10,10,50,50 is followed by a more recently-queued command involving the creation of a larger/covering rectangular region having coordinates 0,0,100,100, then the previously-queued command and related data is deemed unnecessary and is removed from the queue.
Table 1 below lists an exemplary set of protocol commands provided in one embodiment of the invention.
Protocol Command Purpose
Initializes the protocol connection
Instructs the client agent 118 to form a mouse pointer
Instructs the client agent 118 to hide a mouse pointer
Instructs the client agent 118 to change the mode of output text
Instructs the client agent 118 to change the foreground color of
Instructs the client agent 118 to change
the background color of the output text
Instructs the client agent 118 to update the current brush setting
using a new solid color
Instructs the client agent 118 to change the brush pattern
Informs the client agent 118 that the current region is not
Informs the client agent 118 that the current region is clipped
along one side
Informs the client agent 118 that the current region is clipped
along more than one side
Instructs the client agent 118 to change its color palette
Instructs the client agent 118 to output text, represented as one
or more screen glyphs
Instructs the client agent 118 to draw a simple or complex path,
such as a line or bezier curve
Instructs the client agent 118 to draw a particular bitmapped
Instructs the client agent 118 to draw the entire bitmap
Instructs the client agent 118 to draw a portion of the bitmap
Instructs the client agent 118 to copy a particular portion of the
drawing surface to another of the same surface
Instructs the client agent 118 to fill a target region using an
Instructs the client agent 118 to fill a CLIPPING target region
Instructs the client agent 118 to fill a COLOR target region
using a color that is different from the color last used
Instructs the client agent 118 to fill a COLOR_NO_CLIP target
region using a color that is different from the color last used,
Sends data representing a new object to the client agent 118 for
storage in the client's cache
Sends partial data of a new object to the INCOMPLETE client
agent 118 for storage in the client's cache
Sends additional data relating to a previously transmitted
incomplete object and indicates that the object is complete
Sends additional data relating to a INCMPLT previously
transmitted incomplete object and indicates that more data will
Instructs the client agent 118 to write its OBJECTS object
cache, or a portion of its object cache to persistent storage
Instructs the client agent 118 to purge its MEMORY_CACHE
object cache, or a portion of its object cache
Instructs the client agent 118 to read data OBJECT into its
cache from persistent storage
Instructs the client agent 118 to start a timer and write the
current time tick into an array element
Instructs the client agent 118 to determine the difference
between the time tick recorded by the START_STOPWATCH
command and the current time tick
Instructs the client agent 118 to save a rectangular screen area
and to associate it with a particular identifier
Instructs the client agent 118 to restore a SCREEN_BITS
particular rectangular screen area and to then discard the screen
Instructs the client agent 118 to discard specified data
Instructs the client agent 118 to form an off-screen surface
compatible with the current session color depth, associate the
surface with an identifier, and associate attributes to the surface
Instructs the client agent 118 to delete a particular off-screen
Instructs the client agent 118 to switch from a current
destination surface to another specified surface
Informs the server agent 160 of an error in a particular off-
screen surface that makes it unavailable
Informs the server agent 160 that a particular off-screen surface
has become dirty, i.e., that it is not the same as a duplicate
surface accessible to the server agent 160
Instructs the server agent 160 to update KEYS its list of fuzzy
Informs the server agent 160 that objects have been read into
the client's cache
Informs the server agent 160 that objects have not been read
into the client's cache C
Sends the time differential between the RESULT start/stop
stopwatch commands to the server agent 160
Informs the server agent 160 that the protocol connection has
In one embodiment, the protocol commands and associated graphical data are configured to maximize the repetition of information in the protocol packet stream created by the protocol driver. In one aspect of the invention, the graphical data is first encoded (step 204) prior to subjecting the encoded graphical data and/or protocol commands to a compression algorithm (step 206) that takes advantage of the repetitive nature of the protocol stream (e.g., a Lempel-Ziv compression protocol may be used). In one embodiment and upon the first instance of the graphical data in the client-server session, an indicia of the graphical data is transmitted together with the corresponding encoded data in the protocol stream so as to reduce the amount of data transferred via the network 140 for future instances of the graphical data occurring during the same client-server session. The compressed data is subsequently bundled into the transport protocol packet and transmitted to the client node 110 (step 208) for decompression and interpretation.
In one illustrative embodiment, the application program 158 executes a GDI command that instructs the server's operating system 156 to draw a bitmap. The server agent 160 intercepts this GDI command (step 202) and issues a BITBLT command to the client agent 118, which instructs the client agent 118 to display the application program's output data on the client's display 128. Prior to issuing the BITBLT command, the server agent 160 encodes the output data (step 204) and includes the encoded data, together with the BITBLT command in the compressed protocol packet stream transmitted to the client agent 118. In one embodiment, the data describing the bitmap 410 (
In addition to encoding bitmaps or other discrete graphical data elements based on their internal data redundancies, the invention also incorporates caching techniques that further improve the encoding and compression of graphical data that has been previously encountered during a particular client-server session. For example, when a particular bitmap is first encountered during a client-server session it is encoded as described above. For subsequent transmissions of the bitmap during the same client-server session, indicia of the bitmap can be transmitted rather than the relatively larger, encoded bitmap. In one embodiment, the indicia correspond to a location within the client's volatile memory that contains the previously transmitted bitmap.
More particularly and with reference to
Assuming that the protocol command (e.g., BITBLT) and associated graphical data are not superceded by later-issued commands (as described previously), the server agent 160 dequeues (step 360) the BITBLT command and the unique key 412 from the queue and searches the table 411 stored in the server's volatile memory 154 to determine if the unique key 412 had been previously stored therein. If the unique key 412 is located within the table 411, then the server agent 160 determines that the encoded data 414 has been previously processed by the server 150. The server agent 160 then searches the table 411 for an index or handle 416 associated with the encoded bitmap 414. If the index 416 is located, then the server agent 160 determines that the encoded bitmap 414 has not only been processed by the server 150, but that it has also been previously transmitted to the client agent 118. If the index 416 is not located within the table 411, then the server agent 160 determines that the encoded bitmap 414 has not been previously transmitted.
In one embodiment, the index 416 identifies a particular location within the client's volatile memory 114, which stores the previously-transmitted encoded bitmap 414. In one embodiment, the client agent 118 provides a snapshot of its available volatile memory 114 to the server agent 160 upon initiating a client-server session with the server agent 160. Once the session is established, the server agent 160 thereafter controls the allocation of the client's available volatile memory 114. In this manner, the server agent 160 is able to allocate the client's volatile memory 114 to the encoded bitmap 414 and to maintain the index 416 as a pointer or reference to the allocated memory and as an identifier of the previously-transmitted encoded bitmap 414.
If the index 416 is located within the server's table 411, the server agent 160 obtains the index 416 (step 370), compresses the index 416 together with the related protocol command (step 380), and transmits the compressed data in a packet directed to the client agent 118 via the network 140. In one embodiment, the index 416 is shorter in length (e.g., 16 bits) than the unique key 412 (e.g., 64 bits), and consequently consumes less bandwidth when transmitted. Upon receipt and decompression of the compressed packet, the client agent 118 accesses the particular memory location in the client's volatile memory 114 that is specified by the index 416 in order to obtain the appropriate graphical data 410. The client agent 118 subsequently manipulates the obtained graphical data 410 in accordance with the command which accompanied the index 416 in the compressed packet. In this manner, the invention avoids having to retransmit relatively large bitmap or other graphical data that have been previously encountered during a particular client-server session.
If the index 416 is not located within the server's table 411 (and thus has not been previously transmitted to the client agent 118), the server agent 160 allocates a portion of the client's volatile memory 114 for the bitmap 410 and forms the index 416 that identifies the memory location. The server agent 160 then stores the encoded bitmap 414 (corresponding to the bitmap 410), the newly-formed index 416, and the unique key 412 associated with the encoded bitmap 414 in the server's table 411. The server agent 160 subsequently compresses the encoded bitmap 414, the index 416, and related command and transmits the compressed data in a packet directed to the client agent 118. Upon receipt and decompression, the client agent 118 decodes the encoded data 414 and stores the resulting bitmap 410 in the memory location identified by the index 416 and manipulates the resulting bitmap 410 in accordance with the accompanying command. Alternatively, the client agent 118 stores the encoded data 414 in the memory location identified by the index 416 and only decodes the encoded data 414 when necessary to further manipulate the resulting bitmap 410. In this manner, client memory is more efficiently used.
Although the invention has heretofore been described with the server agent 160 controlling the client's memory allocation and index formation, it is also possible that the client agent 118 perform these functions. For example, when a bitmap is first encountered during a client-server session and is encoded, compressed, and transmitted to the client agent 118 (with the accompanying command, but without the index 416), the client agent 118 can determine the particular location within its volatile memory 114 to store the bitmap 410. In this embodiment, the client agent 118 forms an index (not shown) that uniquely identifies the encoded bitmap and its corresponding memory location and transmits the index to the server agent 160, which stores the index in the appropriate location in the server's table 411.
Continuing with the embodiment in which the server agent 160 controls the allocation of the client's volatile memory 114, the client agent 118 maintains a circular queue 421 of all indexes specified by the server agent 160 in accordance with queue maintenance instructions also received therefrom. For example, the server agent 160 can direct the client agent 118 to disassociate a predetermined block of indexes 417 (e.g., corresponding to 128K of memory) from their respective graphical data in order to accommodate newly-encountered graphical objects when the client's volatile memory 114 is otherwise fully subscribed. Consequently, a subsequent new graphical object acquires a previously-used index.
The server agent 160 also determines, according to predetermined criteria, whether to purge or save each graphical object affected by the reassignment of the indexes. The server agent 160 communicates this decision to the client agent 118. For example, if the server agent 160 is reassigning indexes 0, 1, 2, and 3, and determines that the graphical objects associated with handles 0 and 2 are to be deleted or moved into the client's persistent storage 120 and the objects associated with handles 1 and 3 are to be saved, the server agent 160 thus instructs the client agent 118 to delete (or move) 0 and 2, and save 1 and 3. The client agent 118 then moves the index of each saved object to the tail of the circular queue. At least one index remains available at the queue tail for assignment to the moved object; that is, in one embodiment the head and tail of the queue do not point to the same index in the circular queue. Consequently, a moved object is both stored in a different physical memory location and associated with a different index.
The maintenance actions performed on the client's circular queue 421 will typically become more frequent as the client-server session continues and the limited volatile memory 114 of the client 110 becomes full of previously-transmitted bitmaps 410 or other graphical objects. Accordingly, the number of previously-transmitted graphical objects 422 that are moved into the client's persistent storage 120 from volatile memory 114 increases. The present invention further enhances the caching technique described above to search for indicia (see “fuzzy key” below) of previously-transmitted graphical objects 422 that may be stored in the client's persistent storage 120 when an index to that graphical object 422 is not found in the server's table 411. If such indicia is found, then there is a strong likelihood that the previously-transmitted graphical object 422 is still locally resident and accessible to the client agent 118 and that the graphical object 422 may not therefore need to be retransmitted from the server agent 150 to the client agent 110. It is important to note that locating such indicia provides a strong likelihood, but not a certainty, that the graphical object 422 is locally resident, because other programs or client agents which may share the same persistent storage 120 may have deleted the graphical object 422 from persistent storage 120 (e.g., to make room in the persistent storage for a new object).
More particularly and with reference to
In more detail, when the server agent 160 does not find a fuzzy key 419 in the fuzzy database 418 that is associated with the bitmap 422, the server agent 160 sends an encoded bitmap (corresponding to the bitmap 422) to the client agent 118 with a command that requests the client agent 118 to store the bitmap 422 (or the corresponding encoded bitmap) in the volatile memory 114 (step 530). Also and as described above, the server agent 160 adds the fuzzy key 419 associated with the bitmap 422 to the fuzzy database 418 (step 540). Note that the order of steps 530 and 540 can be reversed, such that the fuzzy database 418 is updated prior to the transmission of the command and the encoded version of the bitmap 422. The client agent 118 subsequently copies the bitmap 422 to a particular location within the client's volatile memory 114 as specified by the server agent 160 (step 560).
In one embodiment, the server agent 160 does not wait for an acknowledgement from the client agent 118 that the encoded bitmap associated with the bitmap 422 has been successfully received and processed. Rather, the server agent 160 assumes that the encoded bitmap has been properly processed and continues to stream protocol commands without awaiting the return of an acknowledgement message from the client agent 118. The server agent 160 does, however, track the number of protocol commands issued since the last acknowledgement read. The server agent 160 blocks further transmission of protocol commands if the number reaches a predefined threshold. The threshold enables the client agent 118 to establish an upper bound on the amount of memory needed to queue incoming protocol commands during recovery mode operation, as described in more detail below.
When a fuzzy key 419 is found in the fuzzy database 420, the server agent 160 sends a command to the client 110 directing the client agent 118 to copy the bitmap 422 associated with the fuzzy key 419 from the client's persistent storage 120 into the client's volatile memory 114 (step 550). This command is accompanied not only by the fuzzy key 419 which uniquely identifies the bitmap 422 of interest, but also by the associated index generated by the server agent 160 (as previously described) in order to identify the specific location within the client's volatile memory 114 that will receive and store the copied bitmap 422. By copying the bitmap 422 into volatile memory 114 rather than just maintaining its location in persistent storage 120, the invention can rapidly respond to several requests to draw the bitmap 422 that are received in quick succession or without much intervening cache activity by copying the bitmap 422 on the occurrence of the first request.
If the bitmap 422 is not currently stored in the persistent storage 120 despite the presence of the corresponding fuzzy key 419 in the fuzzy database 418, then the client agent 118 returns a message to the server agent 160 (in response to the “CACHE_READ_DISK_OBJECT” command previously issued by the server agent 160) indicating that the bitmap 422 is missing. Upon receiving this message, the server agent 160 retransmits the encoded bitmap data associated with the bitmap 422 to the client agent 118 (step 530). Note that the server agent 160 maintains a sequential list of all such CACHE_READ_DISK_OBJECT commands sent to the client agent 118 for which the server agent 160 has not yet received an acknowledgement so that the server agent 160 can properly associate a received acknowledgement with a particular bitmap 422 (so as to properly identify the encoded bitmap that needs to be transmitted).
Meanwhile, the client agent 118 enters a recovery mode in which the client agent 118 continues to read the incoming protocol stream, but does not process it. Instead, the client agent 118 builds a queue for receiving the stream of commands that continue to flow from the server 150 subsequent to the command that failed to find the bitmap 422 in persistent storage 120. The client agent 118 continues to store these commands in this queue, in FIFO (first in, first out) fashion, until the encoded bitmap is received and successfully decoded to produce the bitmap 422. To recognize the missing bitmap 422 in the incoming stream of commands, the client agent 118 looks for graphical data accompanying a recovery marker in the protocol stream. In one embodiment, the recovery marker is a particular pseudo-random number that is XORed with the unique key of the bitmap 422 (i.e., the 64-bit CRC and 32-bit length). The server agent 160 creates the recovery marker upon receiving a no-acknowledgement message from the client agent 118. When the bitmap 422 arrives, the client agent 118 stores it in its volatile memory 114 (step 560) (i.e., where the bitmap 422 would have been stored had the object been initially present in the persistent storage 120) and begins to process the commands in the queue. After processing all of the commands in the queue, the client agent 118 resumes processing the incoming stream of protocol commands coming from the server 150 over the network 140.
An advantage to the above-described recovery process is that it avoids the time delay penalty incurred by those cache systems that flush the commands in the pipeline upon the occurrence of a cache miss. Also, the recovery process causes no disruption to the stream of protocol commands nor any increase in the traffic on the channel by more than a few bytes beyond what the server 150 would have sent to the client 110 had the server agent 160 initially known that the bitmap 422 was not in the persistent storage 120.
In one embodiment, the present invention facilitates the location of the bitmap 422 in the client's persistent storage 120 by storing the bitmap 422 in a file whose file name is formed by encoding the fuzzy key 419 along with least-recently-used and/or least-frequently-used information associated with the bitmap 422. In this manner, the file name itself provides the information desired and thus avoids having to open and read the file. The encoding of the fuzzy key 419 as part of the bitmap's file name also enables the client agent 118 to rapidly extract the fuzzy key information from the file name and send it to the server agent 160 during an initial boot of the client 110 or at the start of the client/server session. The extracted fuzzy keys can be transmitted to the server 150 in an order that, for example, is based on the least-recently-used and/or least-frequently-used information that also forms a part of the file name.
In one embodiment, fuzzy keys 419 are added to and deleted from the fuzzy database 418 in one of three ways. First, at the start-up of the client agent 118 or upon establishing a client-server session with the server agent 160, the client agent 118 sends commands to the server agent 160 to add a fuzzy key 419 to the fuzzy database 418 for each appropriate bitmap 422 in the client's persistent storage 120 (the fuzzy keys accompany the client agent's commands). The client agent 118 can maintain a send list that specifies an order (e.g., based on most-recently-used or most-frequently-used information determined from prior client-server sessions) for sending fuzzy keys at startup to the server 150. The send list can be formed by the client agent 118 upon reading the directory of its persistent storage 120 and keeping a representation of the directory in volatile memory 114. In one aspect of the invention, the client agent 118 extracts fuzzy key information from the file name of the file, which contains the bitmap 422, when preparing the representation of the directory. Second, at predetermined polling intervals, the client agent 118 polls its persistent storage 120 to determine those bitmaps that are currently in its persistent storage 120 (e.g., by searching for file names containing a fuzzy key-like structure) and then sends add and/or delete commands to the server agent 160, as needed to update the server's fuzzy database 418. Third, when the server agent 160 sends a command to the client agent 118 to write a bitmap into the client's persistent storage 120, the server agent 160 generates and stores a corresponding fuzzy key 419 in the fuzzy database 418.
Transmission of bitmap data from the server 150 to the client 110 can take a relatively lengthy period of time, particularly when the network connection between the client and server is relatively slow. In one embodiment, the invention adapts the server's operation to changing network conditions by determining the relative speed of the network 140. For example, by dynamically assessing the bandwidth of the network 140, the server agent 160 can modify the encoding and compression techniques used to process bitmap data in order to reduce bandwidth requirements when transmitting over a slow network. The invention processes columns of bitmap data from left to right and primes the client cache 114 accordingly. As the encoded bitmap 414 is being transmitted, the transmission time can be monitored to determine the performance of the network 140. If the transmission time exceeds a threshold value, the client agent 118 can draw whatever bitmap data has already been stored in its cache 114 and display the remaining portions of the bitmap data either in real-time as it is received in the cache 114 or at predetermined intervals. In this manner, a user of the client will recognize that the client 110 is still operating on the bitmap data and be able to discern that the client 110 is not in a hung or failed/error condition. By approximating the effective network bandwidth to the client 110 and adapting the behavior of the server 150 accordingly, the server agent 160 can, for example, instruct the client agent 118 to use a whole bitmap versus using a whole column mode of display for a particular bitmap.
More particularly, the server agent 160 sends a Start_StopWatch command followed by encoded bitmap data 414 and a Stop_StopWatch command to the client agent 118. The client agent 118 responds to the Start_StopWatch command by reading the current time tick and writing it into a first array element. After the encoded bitmap 414 is received, the client agent 118 responds to the Stop_StopWatch command by reading the current time tick and writing it into a second array element. Comparing the difference in time ticks between the Start_StopWatch and Stop_StopWatch commands provides an estimate of the time that was expended to receive the encoded bitmap data. The resulting time difference is then transmitted back to the server agent 160. This procedure can be repeated to compute a smoothed, moving average from which a reasonable estimate of the effective throughput of encoded bitmap data can be determined. The invention uses the throughput estimate in view of the size of the encoded data for a particular bitmap to decide whether to use whole bitmap mode (e.g., if it is likely to take less than 5 seconds) or use progressive whole columns (e.g., if it is likely to take more than 5 seconds). Note that although this procedure does not result in a 100% accurate throughput estimate, it does provide sufficient granularity to detect fast or slow network connections from which different display modes can be determined. A similar technique can be used to time large screen-to-screen copies at the client end and to build up a similar moving average estimate of the client's performance during these operations, which can then be used to slow a producer thread on the server 150 and prevent an overrun at the source when the server 150 generates data faster than the client 110 can handle it.
In one embodiment, the invention forms objects that are sufficient in size to contain the data of a whole cache column. For example, the maximum object size and the size of the cache can be selected by the server agent 160 upon startup of the client/server session to be a function of the size of the connection window and its color depth, with larger sizes being allocated for the larger screens and depths. Further, current techniques for breaking up a bitmap into columns can result in columns of less than standard width at one or both ends of the bitmap. If these nonstandard columns are too narrow, then they may not be large enough to be retained in the client's persistent storage 120 and thus will not be present at the client 110 during subsequent client/server sessions. The invention avoids this problem by merging the narrow columns with neighboring columns so that only wide columns are used and thus all parts of a splash screen, for example, will be contained in the persistent storage 120. In one embodiment, whole columns are then displayed in two phases. The first phase primes the cache 114 with any missing data and the second phase draws the column from the cache 114, which may involve using more than one cached object. The column is therefore displayed on the display screen 128 in an atomic fashion. As previously discussed, when the network connection is slow, the columns can be displayed atomically, one-by-one, to reassure the user at the client 110 that the system is still working. Alternatively, the whole bitmap can be displayed atomically using a similar technique when a fast network connection is detected.
The moving average estimate technique discussed above can also be used to mitigate longstanding overscroll problems in an application that is being executed on a fast server and that is being displayed on a relatively slow client. The overscroll problem occurs, for example, when a user of the client 110 initiates a plurality of scroll commands such as by clicking on the scroll bar (or dragging the scrollbar) with a mouse 126 a number of times. As the scroll commands are sent to the fast server, the server 150 performs the scrolling operations and returns the associated data faster than the slow client 110 (or slow network 140) displays the results on its display screen 128. Consequently when the user at the client 110 eventually views the desired screen position and wants to stop the scrolling (by stopping the mouse clicks and/or scrollbar dragging), the display will continue to scroll beyond the desired screen position. This overscroll condition occurs because the fast server 150 has already processed the scroll commands and has transmitted the appropriate data back to the client 110, but the data has been queued at the slow client and has not yet been entirely processed.
The invention mitigates this overscroll problem by reducing its frequency of occurrence. More particularly, the invention periodically times selected scroll events at the server 150 (by scrolling the frame buffer) and at the client 110 (via the StopWatch commands discussed above) to compute a moving average estimate of their respective speeds. In this manner, the server agent 160 estimates how long a particular scroll event will take (speed times the number of pixels involved) to process at the server 150 and how long the client 110 is expected to take and if the expected processing time at the client 110 is larger than that of the server 150, the server processing is suspended by the appropriate time differential so as to keep the client 110 and server 150 substantially in step. This approach results in many fewer overscrolls due to the time lag between the client 110 and server 150 as compared to the number of overscrolls occurring when this approach is not implemented.
In addition to manipulating bitmap data, the present invention can also remote a text string that is generated on the server 150 to the display 128 of the distant client 110. With reference to
Execution of such procedures cause the display driver 612 to send the appropriate data and protocol commands (such as a subset of those provided in Table 1) to the client agent 118. The protocol commands are received and processed by the client agent 118, which executes procedures 614 that specify the clip region, the bounding rectangle, the text mode, and the glyph associated with each text character to be displayed. Note that the protocol associated with a glyph object encodes not only the glyph bitmap itself, but also its relative positioning information. The server 150 can also send other protocol commands to the client agent 118 that direct the client agent 118 to store the associated glyphs in volatile memory 114 and/or in the persistent storage 120, as previously described. The client agent 118 creates a bitmap corresponding to the size of the bounding rectangle that surrounds the text string and sets the relative positions of each glyph in its proper position within the bounding rectangle. Once the last glyph is received and its position set, the client agent 118 instructs the client operating system 116 to draw/render the bitmap on the display screen 128 of the client 110.
The present invention also supports several different descriptions of paths (a set of lines or curves used as part of the description of a line drawing operation or, in the case of a closed path, to describe a region) including ellipses, beziers, segments, strips, and styles. Segments refer to line segments that, when combined, form a path. Similarly, segments can be further parsed into strips, which are portions of the line segment (corresponding to a series of consecutive pixels) that exhibit the same angle characteristics. The encoding technique used to represent strips, for example, uses relative location information and quantized angles to characterize consecutive strips in particular line segments, which not only minimizes the amount of graphical line data that needs to be encoded but also results in a more repeatable sequence of data that can be more efficiently compressed. The increase in repeatability is particularly enhanced when representing strips that form a shape such as a rectangle or circle. The protocol used for strips also enables the client agent 118 to accurately reproduce a path independently and without knowledge of the algorithm used by the server agent 160 to convert the path into a sequence of pixels.
With reference to the nonlimiting example in
In one embodiment and with reference to
The server agent 160 then parses each line segment 710,720 into strips (712, 714, 716, 718; 722, 724, 726, 728, 730 respectively) (step 830). The length of each strip of a particular segment 710,720 is subsequently stored in a strip length array for further processing (step 840). In one embodiment, the Windows NT algorithm, bLine, is used to break up the path 700 into an array of strips. The NT algorithm determines the quantized angle of each strip and segment so that there is no ambiguity associated with segments or strips that lie on a boundary between quantized angle regions. The NT algorithm vertically flips a line going up so that the line is always going down (the y value keeps increasing) and sets the flag, FL_FLIP_V. Similarly, a line going to the left is flipped horizontally so that it always goes to the right (the x value keeps increasing) and sets the flag, FL_FLIP_H. As a result, the array of strips output by this algorithm fall into a single quadrant. In this particular embodiment, the server agent 160 undoes the flipping operation of the NT algorithm so that the array of strips are flipped back to their previous/original quantized angles to ensure that the endpoint of a previous strip and the starting point of the next consecutive strip coincide so as to avoid having to send the starting coordinate of the next consecutive strip.
The server agent 160 then encodes the beginning coordinate of the path 700 together with the quantized angle data in the line segment array and pixel length data in the strip length array to form a protocol stream (step 850). The encoded data is then subjected to a compression algorithm to take advantage of the data redundancies in the protocol stream (step 860) and the compressed graphical line data is subsequently transmitted to the client agent 118 for decoding and further processing (step 870).
With reference to the exemplary illustration in
Accordingly and in one embodiment, the protocol stream generated by the present invention when processing strips comprises the starting position of the path, an index (if any) corresponding to a location in the client's volatile memory 114 (
The present invention can also be applied to bitmaps that are displayed in an on-screen surface, as well as to bitmaps in an off-screen surface (e.g., that are stored in a video display adapter's memory and/or in a pre-allocated section of the client's volatile memory 114 that will be operated on by a graphics conversion library). Off-screen surfaces are frequently formed by applications, such as Microsoft Word, that write bitmaps to the off-screen surface until the surface is complete, at which time the off-screen surface is displayed on the display screen 128 of the client 110 in final form as an on-screen surface. Therefore, off-screen surfaces frequently provide the source for on-screen surfaces.
In one embodiment and with respect to off-screen surfaces, the client agent 118 informs the server agent 160 as to the amount of local volatile memory 114 available for off-screen surfaces so that the server agent 160 does not form an off-screen surface larger than the client's available memory. The server agent 160 subsequently forms an off-screen surface and transmits it to the client agent 118 with instructions to store the surface in its volatile memory 114. The client agent 118 selects and allocates a particular section of its volatile memory 114 to the surface and stores the surface therein. Each off-screen surface has a set of associated attributes, such as an identifier that uniquely identifies the surface, pixel format, dimensions, drawing attributes, protocol state, etc.
In one embodiment, in order to efficiently use the client's volatile memory 114, the server agent 160 can issue commands to the client agent 118, directing that certain operations be performed to the off-screen surface only when the off-screen surface interacts with the on-screen surface (e.g., when updating the on-screen surface with bitmap data from the off-screen surface). If the only interaction between the on-screen surface and the off-screen surface is a final copy to the screen, then the operations can be targeted to the on-screen surface and thus reduce the amount of bitmap data that must be transmitted over the network 140. Alternatively, the source of the on-screen surface can be restricted to either correspond to an off-screen surface stored remotely at the server 150 or to an off-screen surface stored locally at the client 110. Typical operations to the off-screen surface that are requested by the server agent 160 include updating the bitmap in the off-screen surface with updated text, line, or other data.
In one embodiment and with reference to
Although the present invention has been described with reference to specific details, it is not intended that such details should be regarded as limitations upon the scope of the invention, except as and to the extent that they are included in the accompanying claims.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3596257||Sep 17, 1969||Jul 27, 1971||Burroughs Corp||Method and apparatus for allocating small memory spaces to a computer program|
|US4013828||Feb 20, 1976||Mar 22, 1977||Bell Telephone Laboratories, Incorporated||Method and arrangement for reducing the bandwidth and/or time required to transmit a dithered image|
|US4322795||Jan 24, 1980||Mar 30, 1982||Honeywell Information Systems Inc.||Cache memory utilizing selective clearing and least recently used updating|
|US4395754||Aug 22, 1980||Jul 26, 1983||Siemens Aktiengesellschaft||Data processing system having virtual memory addressing|
|US4410916||Jul 9, 1981||Oct 18, 1983||Compression Labs, Inc.||Dual mode facsimile coding system and method|
|US4430712||Nov 27, 1981||Feb 7, 1984||Storage Technology Corporation||Adaptive domain partitioning of cache memory space|
|US4463424||Oct 21, 1983||Jul 31, 1984||International Business Machines Corporation||Method for dynamically allocating LRU/MRU managed memory among concurrent sequential processes|
|US4499499||Dec 29, 1982||Feb 12, 1985||International Business Machines Corporation||Method for identification and compression of facsimile symbols in text processing systems|
|US4503501||Nov 15, 1982||Mar 5, 1985||Storage Technology Corporation||Adaptive domain partitioning of cache memory space|
|US4513444||Jul 18, 1983||Apr 23, 1985||Dainippon Screen Seizo Kabushiki Kaisha||Method of compressing data|
|US4562423||Jun 5, 1984||Dec 31, 1985||Codex Corporation||Data compression|
|US4691281||Apr 13, 1984||Sep 1, 1987||Nec Corporation||Data processing system simultaneously carrying out address translation of a plurality of logical addresses|
|US4701745||Mar 3, 1986||Oct 20, 1987||Ferranti, Plc||Data compression system|
|US4701871||Oct 1, 1984||Oct 20, 1987||Sony Corporation||Signal generator using digital memory|
|US4796003||Jun 28, 1984||Jan 3, 1989||American Telephone And Telegraph Company||Data compaction|
|US4807029||Jun 16, 1987||Feb 21, 1989||Fuji Photo Film Co., Ltd.||Method of reconstructing image from compression-processed image signals|
|US4860247||Feb 18, 1987||Aug 22, 1989||Hitachi, Ltd.||Multi-window control system|
|US4862392||Mar 7, 1986||Aug 29, 1989||Star Technologies, Inc.||Geometry processor for graphics display system|
|US4870662||Dec 1, 1987||Sep 26, 1989||Concord Data Systems, Inc.||System and method for compressing transmitted or stored data|
|US4899149||Feb 28, 1986||Feb 6, 1990||Gary Kahan||Method of and apparatus for decoding Huffman or variable-length coees|
|US4903218||Aug 13, 1987||Feb 20, 1990||Digital Equipment Corporation||Console emulation for a graphics workstation|
|US4905141||Oct 25, 1988||Feb 27, 1990||International Business Machines Corporation||Partitioned cache memory with partition look-aside table (PLAT) for early partition assignment identification|
|US4928247||Apr 20, 1988||May 22, 1990||Digital Equipment Corporation||Method and apparatus for the continuous and asynchronous traversal and processing of graphics data structures|
|US4937036||Mar 21, 1988||Jun 26, 1990||Xerox Corporation||Concurrent display of data from two different display processors and user interface therefore|
|US4949281||Apr 19, 1988||Aug 14, 1990||H. Berthold Ag||Method and apparatus for generating and producing two-dimensional graphic object by polynominal parametric curves|
|US4958303||May 12, 1988||Sep 18, 1990||Digital Equipment Corporation||Apparatus for exchanging pixel data among pixel processors|
|US4992954||Aug 3, 1988||Feb 12, 1991||Hitachi, Ltd.||Method of storing character patterns and character pattern utilization system|
|US5001478||Dec 28, 1989||Mar 19, 1991||International Business Machines Corporation||Method of encoding compressed data|
|US5016009||Jan 13, 1989||May 14, 1991||Stac, Inc.||Data compression apparatus and method|
|US5049881||Jun 18, 1990||Sep 17, 1991||Intersecting Concepts, Inc.||Apparatus and method for very high data rate-compression incorporating lossless data compression and expansion utilizing a hashing technique|
|US5051745||Aug 21, 1990||Sep 24, 1991||Pkware, Inc.||String searcher, and compressor using same|
|US5072412||Mar 25, 1987||Dec 10, 1991||Xerox Corporation||User interface with multiple workspaces for sharing display system objects|
|US5103303||Feb 13, 1991||Apr 7, 1992||Konica Corporation||Multicolor imaging forming apparatus|
|US5115479||Sep 30, 1991||May 19, 1992||Ricoh Company, Ltd.||Image data processing method for compressing an image by approximating curves using a polynomial|
|US5119319||Dec 14, 1989||Jun 2, 1992||Options Unlimited Research Corp.||Full-duplex video communication system|
|US5126739||Nov 27, 1990||Jun 30, 1992||Stac Electronics||Data compression apparatus and method|
|US5140321||Sep 4, 1991||Aug 18, 1992||Prime Computer, Inc.||Data compression/decompression method and apparatus|
|US5155485||Oct 26, 1990||Oct 13, 1992||Sony Corporation||Digital modulating apparatus and digital demodulating apparatus|
|US5161015||Apr 1, 1991||Nov 3, 1992||Zenith Electronics Corporation||System for peaking a video signal with a control signal representative of the perceptual nature of blocks of video pixels|
|US5164727||Apr 30, 1991||Nov 17, 1992||Regents Of The Unversity Of California||Optimal decoding method and apparatus for data acquisition applications of sigma delta modulators|
|US5195034||Oct 21, 1991||Mar 16, 1993||International Business Machines Corporation||Method for quasi-key search within a National Language Support (NLS) data processing system|
|US5231697||Apr 20, 1990||Jul 27, 1993||Kabushiki Kaisha Toshiba||Method and system for determining connection states of straight short vectors representing figure in curve fitting|
|US5233701||Mar 29, 1989||Aug 3, 1993||Nec Corporation||System for managing interprocessor common memory|
|US5241625||Nov 27, 1990||Aug 31, 1993||Farallon Computing, Inc.||Screen image sharing among heterogeneous computers|
|US5269003||Oct 21, 1992||Dec 7, 1993||Apple Computer, Inc.||Memory architecture for storing twisted pixels|
|US5305440||Jun 12, 1992||Apr 19, 1994||International Business Machines Corporation||File extension by clients in a distributed data processing system|
|US5309554||Sep 5, 1991||May 3, 1994||Brother Kogyo Kabushiki Kaisha||Apparatus for converting character outline data into dot data, having means for moving outline segments|
|US5309555||May 14, 1992||May 3, 1994||International Business Machines Corporation||Realtime communication of hand drawn images in a multiprogramming window environment|
|US5315698||Aug 21, 1991||May 24, 1994||Digital Equipment Corporation||Method and apparatus for varying command length in a computer graphics system|
|US5321806||Aug 21, 1991||Jun 14, 1994||Digital Equipment Corporation||Method and apparatus for transmitting graphics command in a computer graphics system|
|US5339411||Oct 22, 1993||Aug 16, 1994||Pitney Bowes Inc.||Method for managing allocation of memory space|
|US5345588||Sep 17, 1992||Sep 6, 1994||Digital Equipment Corporation||Thread private memory storage of multi-thread digital data processors using access descriptors for uniquely identifying copies of data created on an as-needed basis|
|US5347578||Feb 24, 1993||Sep 13, 1994||International Computers Limited||Computer system security|
|US5351129||Mar 24, 1992||Sep 27, 1994||Rgb Technology D/B/A Rgb Spectrum||Video multiplexor-encoder and decoder-converter|
|US5357623||Nov 1, 1991||Oct 18, 1994||International Business Machines Corporation||Dynamic cache partitioning by modified steepest descent|
|US5384568||Dec 2, 1993||Jan 24, 1995||Bell Communications Research, Inc.||Data compression|
|US5390318||May 16, 1994||Feb 14, 1995||Digital Equipment Corporation||Managing the fetching and replacement of cache entries associated with a file system|
|US5394531||Nov 18, 1991||Feb 28, 1995||International Business Machines Corporation||Dynamic storage allocation system for a prioritized cache|
|US5406279||Sep 2, 1992||Apr 11, 1995||Cirrus Logic, Inc.||General purpose, hash-based technique for single-pass lossless data compression|
|US5414457||Dec 26, 1991||May 9, 1995||Canon Kabushiki Kaisha||Multi-media communication apparatus|
|US5414704||Apr 5, 1994||May 9, 1995||Digital Equipment Corporation||Address lookup in packet data communications link, using hashing and content-addressable memory|
|US5426752||Apr 25, 1994||Jun 20, 1995||Hitachi, Ltd.||Method for allocating real pages to virtual pages having different page sizes therefrom|
|US5434992||Sep 4, 1992||Jul 18, 1995||International Business Machines Corporation||Method and means for dynamically partitioning cache into a global and data type subcache hierarchy from a real time reference trace|
|US5442771||Nov 26, 1993||Aug 15, 1995||Prodigy Services Company||Method for storing data in an interactive computer network|
|US5455576||Feb 7, 1994||Oct 3, 1995||Hewlett Packard Corporation||Apparatus and methods for Lempel Ziv data compression with improved management of multiple dictionaries in content addressable memory|
|US5469540||Jan 27, 1993||Nov 21, 1995||Apple Computer, Inc.||Method and apparatus for generating and displaying multiple simultaneously-active windows|
|US5473742||Feb 22, 1994||Dec 5, 1995||Paragraph International||Method and apparatus for representing image data using polynomial approximation method and iterative transformation-reparametrization technique|
|US5485460||Aug 19, 1994||Jan 16, 1996||Microsoft Corporation||System and method for running multiple incompatible network protocol stacks|
|US5491808||Sep 30, 1992||Feb 13, 1996||Conner Peripherals, Inc.||Method for tracking memory allocation in network file server|
|US5515111||Jan 28, 1994||May 7, 1996||Age Computational Graphics, Ltd.||Method and system for producing enhanced output signals|
|US5515508||Dec 17, 1993||May 7, 1996||Taligent, Inc.||Client server system and method of operation including a dynamically configurable protocol stack|
|US5521597||Aug 2, 1993||May 28, 1996||Mircosoft Corporation||Data compression for network transport|
|US5537551||Nov 18, 1992||Jul 16, 1996||Denenberg; Jeffrey N.||Data compression method for use in a computerized informational and transactional network|
|US5537635||Apr 4, 1994||Jul 16, 1996||International Business Machines Corporation||Method and system for assignment of reclaim vectors in a partitioned cache with a virtual minimum partition size|
|US5561786||Aug 24, 1994||Oct 1, 1996||Microsoft Corporation||Computer method and system for allocating and freeing memory utilizing segmenting and free block lists|
|US5566288||Sep 2, 1994||Oct 15, 1996||Caterpillar Inc.||System and method for automatically fitting a B-spline curve to a set of data points|
|US5572206||Jul 6, 1994||Nov 5, 1996||Microsoft Corporation||Data compression method and system|
|US5579469||Oct 19, 1995||Nov 26, 1996||Lucent Technologies Inc.||Global user interface|
|US5579507||Aug 5, 1993||Nov 26, 1996||Hitachi, Ltd.||Data transfer control of virtual storage supported by three-level hierarchical storage|
|US5588138||Feb 8, 1996||Dec 24, 1996||Gestalt Technologies, Incorporated||Dynamic partitioning of memory into central and peripheral subregions|
|US5651136||Jun 6, 1995||Jul 22, 1997||International Business Machines Corporation||System and method for increasing cache efficiency through optimized data allocation|
|US5652854||Jun 6, 1995||Jul 29, 1997||Novell, Inc.||Method and apparatus for mapping page table trees into virtual address space for address translation|
|US5682486||Mar 14, 1995||Oct 28, 1997||International Business Machines Corporation||Video display and control of multiple graphical interfaces|
|US5682488||Apr 3, 1995||Oct 28, 1997||International Business Machine Corp.||Variable computer icon for single control of complex software functions executed on a data processing system|
|US5717893||May 20, 1991||Feb 10, 1998||International Business Machines Corporation||Method for managing a cache hierarchy having a least recently used (LRU) global cache and a plurality of LRU destaging local caches containing counterpart datatype partitions|
|US5729710||Jun 22, 1994||Mar 17, 1998||International Business Machines Corporation||Method and apparatus for management of mapped and unmapped regions of memory in a microkernel data processing system|
|US5740171||Mar 28, 1996||Apr 14, 1998||Cisco Systems, Inc.||Address translation mechanism for a high-performance network switch|
|US5757795||May 20, 1996||May 26, 1998||Compaq Computer Corporation||Method and apparatus for hashing addresses in a network switch|
|US5757925||Jul 23, 1996||May 26, 1998||Faybishenko; Yaroslav||Secure platform independent cross-platform remote execution computer system and method|
|US5761734||Aug 13, 1996||Jun 2, 1998||International Business Machines Corporation||Token-based serialisation of instructions in a multiprocessor system|
|US5767849||Aug 18, 1995||Jun 16, 1998||International Business Machines Corporation||Personality neutral window management subsystem|
|US5771034||May 5, 1997||Jun 23, 1998||Microsoft Corporation||Font format|
|US5771383||Feb 13, 1997||Jun 23, 1998||International Business Machines Corp.||Shared memory support method and apparatus for a microkernel data processing system|
|US5784070||Jun 3, 1997||Jul 21, 1998||Sony Corporation||Method and apparatus for producing a polygonal image representation through operation of plotting commands on image data|
|US5802281||Sep 7, 1994||Sep 1, 1998||Rsi Systems, Inc.||Peripheral audio/video communication system that interfaces with a host computer and determines format of coded audio/video signals|
|US5826027||Oct 11, 1995||Oct 20, 1998||Citrix Systems, Inc.||Method for supporting an extensible and dynamically bindable protocol stack in a distrubited process system|
|US5835959||Dec 1, 1995||Nov 10, 1998||Sand Technology Systems International, Inc.||Memory management system and method using dual indexing structures|
|US5850632||Sep 6, 1996||Dec 15, 1998||Texas Instruments Incorporated||Memory access controller utilizing cache memory to store configuration information|
|US5864678||May 8, 1996||Jan 26, 1999||Apple Computer, Inc.||System for detecting and reporting data flow imbalance between computers using grab rate outflow rate arrival rate and play rate|
|US5864711||Jul 5, 1995||Jan 26, 1999||Microsoft Corporation||System for determining more accurate translation between first and second translator, and providing translated data to second computer if first translator is more accurate|
|US5874960||Jul 5, 1995||Feb 23, 1999||Microsoft Corporation||Method and system for sharing applications between computer systems|
|US5877757||May 23, 1997||Mar 2, 1999||International Business Machines Corporation||Method and system for providing user help information in network applications|
|US5883640||Aug 15, 1996||Mar 16, 1999||Hsieh; Paul||Computing apparatus and operating method using string caching to improve graphics performance|
|US5886707||Jun 27, 1996||Mar 23, 1999||Berg; David A.||Method for real-time on-demand interactive graphic communication for computer networks|
|US5913230||Mar 6, 1997||Jun 15, 1999||Richardson; John J.||Object and method for providing efficient multi-user access to shared operating system kernal code using instancing|
|US5918019||Jul 29, 1996||Jun 29, 1999||Cisco Technology, Inc.||Virtual dial-up protocol for network communication|
|US5930493||Jun 7, 1995||Jul 27, 1999||International Business Machines Corporation||Multimedia server system and method for communicating multimedia information|
|US5936940||Aug 22, 1996||Aug 10, 1999||International Business Machines Corporation||Adaptive rate-based congestion control in packet networks|
|US5944791||Oct 4, 1996||Aug 31, 1999||Contigo Software Llc||Collaborative web browser|
|US5949975||Mar 12, 1997||Sep 7, 1999||Microsoft Corp.||Method and system for negotiating capabilities when sharing an application program with multiple computer systems|
|US5961588||Feb 21, 1997||Oct 5, 1999||Alcatel Usa Sourcing, L.P.||Handling of commands passed between the server and client stations of a telecommunications system|
|US5968132||Jan 27, 1997||Oct 19, 1999||Fujitsu Limited||Image data communicating apparatus and a communication data quantity adjusting method used in an image data communication system|
|US5983190||May 19, 1997||Nov 9, 1999||Microsoft Corporation||Client server animation system for managing interactive user interface characters|
|US5999179||Nov 17, 1997||Dec 7, 1999||Fujitsu Limited||Platform independent computer network management client|
|US5999950||Aug 11, 1997||Dec 7, 1999||Webtv Networks, Inc.||Japanese text input method using a keyboard with only base kana characters|
|US6014133||Mar 11, 1997||Jan 11, 2000||Seiko Epson Corporation||Data transmitter/receiver apparatus, data transmitter, data receiver, and data compression method|
|US6016535||Oct 11, 1995||Jan 18, 2000||Citrix Systems, Inc.||Method for dynamically and efficiently caching objects by subdividing cache memory blocks into equally-sized sub-blocks|
|US6023749||Jan 25, 1999||Feb 8, 2000||Citrix Systems, Inc.||Object and method for providing efficient multi-user access to shared operating system kernal code using instancing|
|US6034689||Jun 3, 1996||Mar 7, 2000||Webtv Networks, Inc.||Web browser allowing navigation between hypertext objects using remote control|
|US6052120||Oct 1, 1996||Apr 18, 2000||Diamond Multimedia Systems, Inc.||Method of operating a portable interactive graphics display tablet and communications systems|
|US6057857||Jun 12, 1996||May 2, 2000||Citrix Systems, Inc.||Method for the lossless compression of lines in a distributed computer system|
|US6061714||May 7, 1997||May 9, 2000||International Business Machines Corporation||Persistent cache synchronization and start up system|
|US6061715||Apr 30, 1998||May 9, 2000||Xerox Corporation||Apparatus and method for loading and reloading HTML pages having cacheable and non-cacheable portions|
|US6081623||Oct 11, 1995||Jun 27, 2000||Citrix Systems, Inc.||Method for lossless bandwidth compression of a series of glyphs|
|US6161126||Feb 2, 1999||Dec 12, 2000||Immersion Corporation||Implementing force feedback over the World Wide Web and other computer networks|
|US6163773||May 5, 1998||Dec 19, 2000||International Business Machines Corporation||Data storage system with trained predictive cache management engine|
|US6181711||Oct 10, 1997||Jan 30, 2001||Cisco Systems, Inc.||System and method for transporting a compressed video and data bit stream over a communication channel|
|US6233606||Dec 1, 1998||May 15, 2001||Microsoft Corporation||Automatic cache synchronization|
|US6314452||Aug 31, 1999||Nov 6, 2001||Rtimage, Ltd.||System and method for transmitting a digital image over a communication network|
|US6397259||May 29, 1998||May 28, 2002||Palm, Inc.||Method, system and apparatus for packet minimized communications|
|US6449639||Dec 23, 1998||Sep 10, 2002||Doxio, Inc.||Method and system for client-less viewing of scalable documents displayed using internet imaging protocol commands|
|US6453343||Jan 21, 1999||Sep 17, 2002||International Business Machines Corporation||Methods, systems and computer program products for maintaining a common checkpoint cache for multiple sessions between a single client and server|
|US6489956||Oct 6, 1999||Dec 3, 2002||Sun Microsystems, Inc.||Graphics system having a super-sampled sample buffer with generation of output pixels using selective adjustment of filtering for implementation of display effects|
|US6603470||Nov 4, 1997||Aug 5, 2003||Sun Microsystems, Inc.||Compression of surface normals in three-dimensional graphics data|
|US6639593||Jul 31, 1998||Oct 28, 2003||Adobe Systems, Incorporated||Converting bitmap objects to polygons|
|US6657954||Mar 31, 1999||Dec 2, 2003||International Business Machines Corporation||Adapting receiver thresholds to improve rate-based flow control|
|US6731600||Feb 8, 1999||May 4, 2004||Realnetworks, Inc.||System and method for determining network conditions|
|US6801499||Dec 14, 1999||Oct 5, 2004||Texas Instruments Incorporated||Diversity schemes for packet communications|
|US7028025||May 25, 2001||Apr 11, 2006||Citrix Sytems, Inc.||Method and system for efficiently reducing graphical display data for transmission over a low bandwidth transport protocol mechanism|
|US7127525||May 25, 2001||Oct 24, 2006||Citrix Systems, Inc.||Reducing the amount of graphical line data transmitted via a low bandwidth transport protocol mechanism|
|US20010019630||Mar 31, 1999||Sep 6, 2001||America Online, Inc.||Method for transferring and displaying compressed images|
|US20020029285||May 25, 2001||Mar 7, 2002||Henry Collins||Adapting graphical data, processing activity to changing network conditions|
|US20020035596||May 25, 2001||Mar 21, 2002||Ruiguo Yang||Remote control of a client's off-screen surface|
|US20020049833||May 4, 1998||Apr 25, 2002||Dan Kikinis||Tailoring data and transmission protocol for efficient interactive data transactions over wide-area networks|
|US20020073061||May 25, 2001||Jun 13, 2002||Henry Collins||Method and system for efficiently reducing graphical display data for transmission over a low bandwidth transport protocol mechanism|
|US20030046432||May 25, 2001||Mar 6, 2003||Paul Coleman||Reducing the amount of graphical line data transmitted via a low bandwidth transport protocol mechanism|
|US20030084052||Feb 4, 1994||May 1, 2003||Richard E. Peterson||Computerized information retrieval system|
|USRE32632||Dec 20, 1985||Mar 29, 1988||Apple Computer, Inc.||Display system|
|EP0283735A2||Feb 22, 1988||Sep 28, 1988||Hayes Microcomputer Products, Inc.||Adaptive data compression method and apparatus|
|EP0684582B1||Apr 1, 1995||Oct 6, 1999||Agfa Corporation||Method and apparatus for reconstruction of character outlines|
|EP0684583B1||Apr 4, 1995||Oct 6, 1999||Agfa Corporation||Method for construction of a scaleable font database|
|EP0691628A2||Jul 5, 1995||Jan 10, 1996||Microsoft Corporation||Data compression method and system|
|EP0734144A2||Mar 4, 1996||Sep 25, 1996||Siemens Aktiengesellschaft||Method and apparatus for determination of user charges in a subscriber apparatus|
|EP0739140A2||Apr 17, 1996||Oct 23, 1996||Sun Microsystems Inc.||Encoder for an end-to-end scalable video delivery system|
|EP0818753B1||Jul 9, 1997||Oct 1, 2003||Samsung Electronics Co., Ltd.||Contour decoding apparatus|
|EP0836145B1||Oct 1, 1997||Jun 7, 2006||AT&T Corp.||Method for transferring and displaying data pages on a data network|
|GB2136171A||Title not available|
|GB2323946B||Title not available|
|1||"A New Source Coding Method Based on LZW Adoptind the Least Recently Used Deletion Heuristic" by Hayashi, Kubo, Yamazoto & Sasase; (pp. 190-193); May 19, 1993; University of Victoria.|
|2||"Creating a Wider Bus Using Caching Techniques" by Citron & Rudolph; Institute of Computer Science, Hebrew University; Jerusalem, Israel; .COPYRGT.1995 IEEE, pp. 90-99.|
|3||"The Norton Utilities Version 5.0 Disk Explorer", Copyright .COPYRGT. 1990 Peter Norton Computing, Inc.; (pp. 28-54); Santa Monica, CA.|
|4||"Word Based Data Compression Schemes" by Bar-Ness & Peckham, .COPYRGT. 1989 (pp. 300-303), The Center for Communication and Signal Processing Research Department of Electrical Engineering, NJIT Newark, NJ.|
|5||Cate et al, "Combining the concepts of compression and caching for a two-level file system", ACM 1991, pp. 200-209.|
|6||European Search Report, Application EP03 00 5805.1, 3pgs.|
|7||European Search Report, Application No. EP03 00 5803.6, 3 pgs.|
|8||European Search Report, Application No. EP03 00 5804.4, 3 pgs.|
|9||European Search Report, Application No. EP03 00 5806.9, 3 pgs.|
|10||First Examination Report, 549/KOL/2006 dated Dec. 22, 2010.|
|11||First Examination Report, 550/KOL/2006 dated Dec. 22, 2010.|
|12||Freeman, H., "A Scheme for the Efficient Encoding of Graphical Data for Communication and Information Processing", Proceedings of the Electronics Congress-Advance in Electronics, Rassegna, IT, pp. 340-348, Mar. 24, 1969.|
|13||Freeman, H., "A Scheme for the Efficient Encoding of Graphical Data for Communication and Information Processing", Proceedings of the Electronics Congress—Advance in Electronics, Rassegna, IT, pp. 340-348, Mar. 24, 1969.|
|14||Freeman, H., "Computer Processing of Line-Drawing Images", ACM Computing Surveys, New York, NY, US, vol. 6, No. 1, pp. 57-97, Mar. 1974.|
|15||Hammami, "Cache memories, fuzzy localities and fuzzy working sets", IEEE 1996, pp. 535-540.|
|16||IBM Technical Disclosure Bulletin, "Mechanisms for Finding Substitute Fonts in a Conference-Enabled X Windows Application", vol. 41, No. 1, Jan. 1998, pp. 137-142.|
|17||IBM Technical Disclosure Bulletin, "Remote Desktop Environments Reflected in Local Desktop Windows", vol. 36 No. 3, Mar. 1993, pp. 421-426.|
|18||Murray, et al., "1.5 Move-to-Front Coding", 4 pp. 16-19, 1994.|
|19||Patents on data compression algorithms; available web site: www.iapr-ic.dimi.uniud.it; accessed on: Mar. 31, 1998; p. 1-10.|
|20||WebTV, "WebTV System Guide", http://developer.webtv.net/authoring/sysgde/sysgde.html, Sep. 13, 2000, pp. 1-12.|
|21||Yazici et al, "An index structure for fuzzy database", IEEE 1996, pp. 1375-1381.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US20130019024 *||Jul 14, 2011||Jan 17, 2013||Qualcomm Incorporatd||Wireless 3d streaming server|
|U.S. Classification||707/636, 709/217|
|International Classification||G06F7/00, G06F17/30, H04L29/06, H04L29/08, G06F3/14, G06F17/00, G06F3/153|
|Cooperative Classification||Y10S707/99945, Y10S707/99933, Y10S707/99932, G06F17/30905, H04L67/36, H04L69/04, H04L69/329, H04L67/38, G06F3/1454, H04L67/42, H04L67/2876, H04L67/2842, H04L67/1002, H04L67/289, H04L67/2828|
|European Classification||H04L29/08N27L, H04L29/08A7, G06F3/14T, H04L29/08N9A, H04L29/06C5, H04L29/08N35, G06F17/30W9V, H04L29/08N27S, H04L29/06C4, H04L29/06C8|
|Feb 9, 2009||AS||Assignment|
Owner name: CITRIX SYSTEMS, INC., FLORIDA
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:COLLINS, HENRY;REEL/FRAME:022224/0450
Effective date: 20010703
|Jul 1, 2015||FPAY||Fee payment|
Year of fee payment: 4